The invention relates to a method for electronically evaluating a dialog between at least two persons comprising receiving audio data, analyzing the audio data to determine the reparation of utterances of the at least two persons in the course of the dialog and comparing the results of the analysis with predetermined communication patterns.
Legal claims defining the scope of protection, as filed with the USPTO.
1. A method for automatically, in particular electronically, evaluating a dialogue between at least two persons, comprising: receiving a video data stream showing the at least two persons at least partially, and analyzing the video data stream to determine whether a visual interaction between the at least two persons occurs during the dialogue, comprising determining whether the at least two persons look at each other by detecting the position of at least a portion of an iris of an eye of each of the at least two persons, and analyzing movement of the at least a portion of the iris of the eye for each of the at least two persons to determine whether the movements of the eyes of the at least two persons are in a synchronized state with each other, wherein analyzing the movements includes analyzing the movement of the at least a portion of the iris into one of the six directions of the eye: upper left, upper right, left, right, downwards left and downwards right, to determine if the movements of the eyes of the at least two persons are in the synchronized state.
2. The method according to claim 1 , wherein analyzing the movement of at least a portion of the eyes further comprises: determining whether the movements of the eye are in the synchronized state in at least one predetermined interaction phase.
3. The method according to claim 1 , wherein analyzing the video data stream further comprises: identifying visual features in a picture of the video data stream to determine the zone of the picture showing at least a part of the face of at least one of the two persons, isolating stripes, in particular vertical stripes, in the zone, and searching for eye zones in the stripes if more than one stripe has been isolated.
4. The method according to claim 3 , wherein the step of isolating stripes further comprises: identifying contours in the stripes by transferring the stripes in a grey-scale format and searching candidate centers of the determined contours by searching for circles susceptible to correspond to the iris by weighting the grey scale intensity in the video data stream in the contours with a predetermined color.
5. The method according to claim 3 , further comprising: identifying the position of at least a portion of the iris in an eye zone, by searching for circles susceptible to correspond to the iris by weighting the grey scale video data stream in the contours with a predetermined color.
6. The method according to claim 5 , wherein the step of searching for circles susceptible to correspond to the iris is repeated to identify the position of the iris in an eye zone, unless no iris position is detectable.
7. The method according to claim 1 , wherein the video data stream comprises a plurality of data streams from different cameras located at different positions with respect to the persons participating in the dialogue.
8. The method according to claim 7 , wherein each camera captures only one person.
9. A system for automatically, in particular electronically, evaluating a dialogue between at least two persons, comprising: at least one video camera for capturing a video data stream showing the at least two persons at least partially and during the dialogue, and a processing device coupled to the at least one video camera and configured to determine whether a visual interaction between the at least two persons occurs during the dialogue, wherein the processing device is further configured to detect a position of at least an eye of a person in the video data stream comprising a plurality of subsequent pictures by identifying visual features in a picture of the plurality of subsequent pictures, and to determine a zone of the picture showing at least a part of a face of the person by isolating vertical stripes in the zone, and by searching for one or more eye zones in the vertical stripes if more than one of the vertical stripes has been isolated, wherein the processing device is further configured to detect a position of at least an eye for each of the at least two persons, and to identify whether movements of the eyes of the at least two persons are in a synchronized state with one and other, and wherein for each of the eyes the processing device is configured to determine the movement of the eye into one of the following six directions of the eye: upper left, upper right, left, right, downwards left and downwards right, to determine if the movements of the eyes of the at least two persons are in the synchronized state.
10. The system according to claim 9 , wherein for each person participating in the dialogue one video camera is provided, and wherein the video cameras are arranged such that each camera captures only one person.
11. The method according to claim 1 , further comprising a computer program product, wherein one or more computer readable media has computer-executable instructions for performing the steps of the method.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
June 25, 2008
October 14, 2014
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.